• The Transactions of the Korea Information Processing Society
• /
• v.7 no.12
• /
• pp.3821-3828
• /
• 2000

Many software projects collect grouped failure data (failures in some failure interval or in variable time interval) rather than individual failure times or failure count data during the testing or operational phase. This paper presents the neural network (NN) modeling that is dble to predict cumulative failures in the variable future time for grouped failure data. ANN's predictive ability can be affected by what it learns and in its ledming sequence. Eleven training regimes that represents the input-output of NN are considered. The best training regimes dre selected rJdsed on the next' step dvemge reldtive prediction error (AE) and normalized AE (NAE). The suggested NN models are compared with other well-known KN models and statistical software reliability growth models (SHGlvls) in order to evaluate performance, Experimental results show that the NN model with variable time interval information is necessary in order to predict cumulative failures in the variable future time interval.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.4
• /
• pp.765-775
• /
• 2010

For the accelerated failure time (AFT) model a lot of effort has been devoted to develop effective estimation methods. AFT model assumes a linear relationship between the logarithm of event time and covariates. In this paper we propose a semiparametric support vector machine to consider situations where the functional form of the effect of one or more covariates is unknown. The proposed estimating equation can be computed by a quadratic programming and a linear equation. We study the effect of several covariates on a censored response variable with an unknown probability distribution. We also provide a generalized approximate cross-validation method for choosing the hyper-parameters which affect the performance of the proposed approach. The proposed method is evaluated through simulations using the artificial example.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.3
• /
• pp.861-869
• /
• 2006

Burn-in is a widely used method to improve the quality of products or systems after they have been produced. In this paper, the problem of determining optimal burn-in time for a system which performs given mission is considered. It is assumed that the given mission time is not a fixed constant but a random variable which follows an exponential distribution. Assuming that the underlying lifetime distribution of a system has an eventually increasing failure rate function, an upper bound for the optimal burn-in time which maximizes the probability of performing given mission is derived. The obtained result is also applied to an illustrative example.

• Geomechanics and Engineering
• /
• v.34 no.1
• /
• pp.1-15
• /
• 2023

This study conducts uniaxial compression tests on intact and single crack-contained rocks to investigate the time-frequency domain characteristics of acoustic emission (AE) signals monitored during the deformation failure process. A processing approach, short-time Fourier transform (STFT), is performed to obtain the evolution characteristics of time-frequency domain of AE signals. The AE signal modes at different deformation stages of rocks are different. Five modes of AE signal are observed during the cracking process of rocks. The evolution characteristics of time-frequency domain of AE signals processed by STFT can be utilized to evaluate the damage process of rocks. The difference of time-frequency domain characteristics between intact and cracked rocks is comparatively analyzed. The distribution characteristics of frequency changing from a single band-shaped cluster to multiple band-shaped clusters can be regarded as an early warning information of damage and failure of rocks. Meanwhile, the attenuation of frequency enables the exploration of rock failure trends.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.21 no.5
• /
• pp.599-606
• /
• 2018

Failure data of systems in many field can be erroneous, which influences the reliability analysis of the systems. The general form of failure data is right censored data with accurate time information. But due to its nature of data collection in the military field, failure time of one-shot weapon systems can have errors which are related to the maintenance period. So this paper suggests a model that can reduce the error by utilizing interval censored data as an alternative to right censored data in weibull distribution.

• Journal of the Korean Society for Railway
• /
• v.9 no.4 s.35
• /
• pp.419-424
• /
• 2006

In this paper we study on a method to predict and to demonstrate the reliability of the Korea high speed train control system in quantitative point of view. For the prediction of the reliability in train control system which is composed of electronic parts, Relax Software 7.7 automation tool is employed and MIL-HDBK-217 Handbook that is a standard for the prediction of the failure rate in electronic components is used. Mean Time Between Failure (MTBF) is predicted based on the failure rate of the subsystems, State Modeling and Markov Modeling method is used to express a reliability function of the train control system composed by hardware redundancy as a function of time. We propose a Reliability Test which is performed on the level of the subsystems and Failure Report, Analysing, Correction action system which use the test operation data to prove the predicted reliability.

• ETRI Journal
• /
• v.44 no.3
• /
• pp.361-378
• /
• 2022

In 5G, the required target for interruption time during a handover (HO) is 0 ms. However, when a handover failure (HOF) occurs, the interruption time increases significantly to more than hundreds of milliseconds. Therefore, to fulfill the requirement in as many scenarios as possible, we need to minimize HOF rate as close to zero as possible. 3GPP has recently introduced conditional HO (CHO) to improve mobility robustness. In this study, we propose "ZEro handover failure with Unforced and automatic time-to-execute Scaling" (ZEUS) algorithm to optimize HO parameters easily in the CHO. Analysis and simulation results demonstrate that ZEUS can achieve a zero HOF rate without increasing the ping-pong rate. These two metrics are typically used to assess an HO algorithm because there is a tradeoff between them. With the introduction of the CHO, which solves the tradeoff, only these two metrics are insufficient anymore. Therefore, to evaluate the optimality of an HO algorithm, we define a new integrated HO performance metric, mobility-aware average effective spectral efficiency (MASE). The simulation results show that ZEUS provides higher MASE than LTE and other CHO variants.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.18 no.1
• /
• pp.71-78
• /
• 1993

A Baysian approach is proposed is estimating the mission failure rates by criticalities. A mission failure which occurs according to a Poisson process with unknown rate is assumed to be classified as one of the criticality levels with an unknown probability. We employ the Gamma prior for the mission failure rate and the Dirichlet prior for the criticality probabilities. Posterior distributions of the mission rates by criticalities and predictive distributions of the time to failure are derived.

• Management Science and Financial Engineering
• /
• v.7 no.2
• /
• pp.73-90
• /
• 2001

This paper presents the reliability models for redundant systems composed of repairable components whose failure time and repair time distributions are phase-type. It is shown that the distribution of time to system failure is also phase-type. The dependency between components are considered and integrated into the model by the used of the rate adjustment factor. The phase-type representation is constructed for the system through algebraic operations on the parameters of components\` failure time and repair time distributions and the corresponding rate adjustment factors. Types of system structures considered are parallel, k-out-of-N system with load sharing scheme and standby system with operation priority.

• Journal of Communications and Networks
• /
• v.8 no.2
• /
• pp.228-233
• /
• 2006

Survivability is a very important requirement for the deployment of broadband networks because out of service links can affect volumes of traffic even if it is a very short time. And the data paths of broadband networks, which are critical for traffic engineering, are always necessary to be well protected. The procedure of protection or restoration for a path is initiated when failure is detected within the working path. In order to minimize the influence on transmission quality caused by the failure of links and to provide a definite time for the recovery from the failure, the protection switching time (PST) should be carefully considered in the path arrangement. Several researches have been devoted to construct the protection and restoration schemes of data paths over dense wavelength division multiplexing (DWDM) networks, however, there was rare research on the design of data paths with guaranteed protection switching time. In this paper, the PST-guaranteed scheme, which is based on the concept of short leap shared protection (SLSP), for the arrangement of data paths in DWDM networks is proposed. The proposed scheme provides an efficient procedure to determine a just-enough PST-guaranteed backup paths for a working path. In addition to selecting the PST-guaranteed path, the network cost is also considered in a heuristic manner. The experimental results demonstrate that the paths arranged by the proposed scheme can fully meet the desired PST and the required cost of the selected path is competitive with which of the shared path scheme.